Brexit Survey Analysis
This report gives a brief summary of the textual analysis of the submissions to the Brexit Survey by the Leeds Chamber of Commerce.
Summary of Key Points
- A total of 364 comments were posted to this survey.
- Primary school-level educated participants.
- Generally quite positice submissions with anticipation and trust also being common sentiments.
- Main issues raised are of the price of raw materials and stock holding.
Frequency Analysis
The comments had an average of 11 words in each, however almost 150 of the respondants responded with “none”. There was an average Flesch readability score of 71 suggesting readers needed to be educated to at least a UK Grade Level of 6 to understand the comments. As this engagement activity was in survey format, the people making submissions were educated just to primary school level. The very short length of the comments also suggests very little qualitative reponse from the participants.
The most common adjectives, phrases and pairs of words are displayed below. People tend to express their emotions through the adjectives they use, and in this case “raw”, “more”, and “difficult” being used so often relate to the main concern of the increased price and difficulty regarding raw materials after Brexit. 
A network of the most frequent consecutive word pairs (bigrams) is shown below. “increased stock”, “raw materials”, and “supply chain” are the most common word pairs in the dataset. Phrases such as “awaiting outcome”, “direct impact”, and “what’s happening” are also common and suggest pockets of comments which express uncertainty and worry at the impending issue and its effects on their business. The presence of “overseas students” also suggests these worries extend beyond the industries.
Topic Extraction
A plot of words most associated with one of 8 topics are shown below.
Topic model visualisations are split into two sections:
Left - showing topic distances from each other based on the types of words in each,
Right – showing the top 30 words in each topic (red bar) and overall in the dataset (blue bar). I recommend setting the relavance metric to 0.6 to get a more representative list of words in each topic.
This visualisation is interactive, hover over each topic number to view the words in each topic, or select each word to view which topics it appears.
https://nicolednisbett.github.io/Brexit-Survey/#topic=0&lambda=0.62&term=
Sentiment Analysis
The wordcloud below gives the most popular words associated with positive and negative sentiments in the survey. Specific comments which are associated with the most popular sentiments are listed below.
The NRC sentiment lexicon uses categorical scale to measure 2 sentiments (positive and negative), and 8 emotions (anger, anticipation, disgust, trust, joy, sadness, fear, and suprise). Examples of words and comments in these sentiment categories are below. In this debate, the majority of submissions were negative but also categorised as anticipation and positive.
Hover over the plot below to read the content of the comments within each sentiment category.
## [1] 7
##
## anger anticipation disgust fear joy
## 0.06748466 0.16564417 0.01226994 0.12269939 0.03680982
## negative positive sadness surprise trust
## 0.12883436 0.22699387 0.05521472 0.03680982 0.14723926An example of a comment categorised as negative
No plans for staff recruitment in case there is a downturn in demand for services
An example of a comment categorised as anticipation
Waiting to see what happens, continue as normal and wait for the regs to catch up.
An example of a comment categorised as trust
Wait and see what happens. There is no definitive analysis to base any assumptions on.